Optimization , production and scale up of debittered kinnow beverage by α-L-rhamnosidase producing yeast

R E G U L A R A R T I C L E Singh, et al.: Production of debittered kinnow beverage Emir. J. Food Agric ● Vol 27 ● Issue 7 ● 2015 549 benzene styrene resin treatment and β-cyclodextrin treatment (Mongkolkul et al., 2006). These techniques have limitations in altering nutrient composition either through chemical reactions or removal of nutrients, flavor and color etc. Another suitable debittering procedure is the stepwise hydrolysis of naringin by naringinase (Chen et al., 2010). The enzyme naringinase is composed of α-L-rhamnosidase (EC 3.2.1.40) and β-D-glucosidase (EC 3.2.1.21). Naringin (4′,-5,7′-trihydroxyflavonone-7-rhamnoglucoside) is first hydrolyzed by α-L-rhamnosidase activity of naringinase to rhamnose and prunin (one third of the bitterness of naringin) which can be further hydrolyzed into glucose and naringenin by the β-D-glucosidase component of naringinase. The potential application of rhamnosidase is used in the debittering of citrus fruit juices (Busto et al., 2007), manufacture of prunin from naringin, manufacture of L-rhamnose by hydrolysis of natural glycosides containing terminal L-rhamnose, enhancement of wine aromas by enzymatic hydrolysis of terpenyl glycosides containing L-rhamnose, elimination of hesperidin crystals from orange juices, conversion of chloropolysporin B to chloropolysporin C, the derhamnosylation of many L-rhamnose containing steroids for example, diosgene, desglucoruscin, ginsenosides-Rg2, etc. whose derhamnosylated products have their clinical importance (Feng et al., 2005). The nutritional and therapeutic value of kinnow provides ample scope for processing into a value added fermented product with retention of organoleptic properties, nutritional attributes, characteristics sensory properties, flavor, aroma, texture and long shelf life. So, the aim of this work is to optimize and produce debittered kinnow beverage using α-L-rhamnosidase producing yeast Clavispora lusitaniae KF633446. MATERIALS AND METHODS Yeast culture Yeast strain producing rhamnosidase enzyme was isolated from whey beverage and identified as Clavispora lusitaniae (accession numberKF633446) on the basis of morphological, biochemical and 18S rDNA sequence analysis. Screening of juice components for optimized α-Lrhamnosidase production The physical and nutritional conditions were optimized following ‘one-at-a-time’ approach to enhance the yield of α-Lrhamnosidase enzyme in diluted kinnow juice (juice:water; 1:1.5). The effect of percent inoculum concentration (0.25, 0.5, 0.75, 1 and 1.25 v/v), total soluble solids in (12, 13, 14, 15 and 16 °B), incubation time (12, 24, 36, 48 and 60 h) and temperature (15, 20, 25, 30 and 35 °C) on enzyme activity were evaluated. For each parameter optimization, three sets of independent experiments were carried out and the average value was reported. α-L-Rhamnosidase enzyme assay 50 mL of the diluted kinnow juice in Erlenmeyer flask (100 mL) was aerobically cultured at 30±5 °C for 48 h on a rotary shaker (150 rpm). After centrifugation (12,000 × g for 10 min), the supernatant was collected for measurement of rhamnosidase activity. The α-L-rhamnosidase activity (RA) was determined using p-nitrophenyl-α-L-rhamnoside (p-NPR, Sigma) as the substrate (Romero et al., 1985). The reaction mixture consisted of 0.1 mL of 4.8 mM p-NPR solution, plus 0.19 mL of 50 mM citric acid/Na citrate buffer, pH 5.0 and 10 μl of enzyme or buffer (for the blank) and was incubated at 50 °C. Aliquots of 50 μL from the reaction mixture were removed every 2 min and placed into 1.5 mL of 0.5 M NaOH. These aliquots were kept in an ice bath until the absorbance was measured at 400 nm (Rajal et al., 2009). One unit (U) of enzyme activity was defined as the amount of enzyme required to release 1 μmol of p-nitrophenol per minute. Preparation of debittered kinnow beverage A debittered kinnow beverage was prepared under optimized conditions of inoculum concentration, TSS, temperature and incubation time. Extraction of juice Kinnow (Citrus reticulata Blanco) was procured from Department of Fruit Science, PAU, Ludhiana, Punjab, India. Fruits were washed in chlorinated water and then used for the extraction of juice. Juice was extracted aseptically under hygienic conditions by kinnow juice extractor. Preparation of sugar solution The sugar solution was prepared by boiling (500 g) granulated sucrose in one litre of water for 10 min and then allowed to cool at room temperature and stored aseptically in sterilized glass bottles. Inoculum preparation The inoculum was prepared in diluted juice with brix adjusted to (13°B). A loopful culture of 24 h old yeast (Clavispora lusitaniae KF633446) was inoculated in 100 mL diluted kinnow juice in 250 mL Erlenmeyer flask and incubated at 30±5 °C for 24 h to achieve concentration of 105106 cells mL-1. Fermentation, bottling and storage The physico-chemical analysis (pH, % acidity, TSS, brix acid ratio, naringin, limonin and juice yield) of fresh kinnow juice was performed. Juice was diluted in the ratio 1:1.5 with water. Diluted juice was pasteurized at 82 °C for Singh, et al.: Production of debittered kinnow beverage 550 Emir. J. Food Agric ● Vol 27 ● Issue 7 ● 2015 15 sec, cooled and brix adjusted to 13 °B by adding sugar solution followed by inoculation of yeast i.e. 0.75% (v/v). It was incubated for 48 h at 30±5 °C. The beverage was refrigerated for 24 h, siphoned, bottled and stored in refrigerated conditions. Shelf life determination of kinnow beverage Shelf life fermented debittered kinnow beverage, stored at refrigerated temperature (4°C) was studied and evaluated fortnightly for physicochemical, microbiological and sensory qualities. Physiochemical and microbiological analysis of kinnow juice and beverage The total soluble solids and pH of kinnow juice and beverage were determined by using Erma hand refractometer of 0-32 °B (Erma, Tokyo, Japan) and pH meter (ECIL, Hyderabad, type 101; Electronic Corporarion of India Ltd., Hyderbad, India). Total acidity expressed as citric acid was estimated following the procedure of AOAC (1999). Brix-acid ratio was calculated through dividing TSS value by total acidity of the juice and carbonated beverage. Total sugars were estimated by phenol sulphuric acid method (Dubois et al., 1956). Reducing sugars were estimated by the method of Miller (1959). The titration method using 2, 6-dichlorophenol indophenol dye was used to estimate ascorbic acid (AOVC 1996). The total phenolic content (TPC) was determined by spectrophotometry, using gallic acid as a standard, according the method described by Singleton and Rossi (1965). Limonin content was estimated by colorimetric method (Vaks and Lifshitz, 1981) and naringin content was estimated by Davis method (1947). Carbon dioxide volumes in beverage bottles were determined by Zahm and Nagel piercing device (CO2 tester, Zahm and Nagel Co., Inc., Holland, New York, USA) and percent alcohol (v/v) was calculated by spectrophotometric determination method of ethanol (Caputi et al., 1968). Total yeast count was enumerated on GYE agar by serial plate dilution method. Sensory evaluation The organoleptic evaluation of kinnow beverages was done on the basis of appearance, taste, color, aroma, bouquet, body, flavor, astringency and overall acceptability by a panel of judges. Consumer acceptance for the products was evaluated on a nine point “Hedonic scale” (Amerine et al., 1965). Statistical analysis Statistical analysis was done by using CPCS1 software. Standard errors were calculated for all mean values. Differences were considered significant at the p ≤ 0.05 level. RESULTS AND DISCUSSION Screening of juice components for optimization of α-L-rhamnosidase production Effect of percent inoculum concentration on α-Lrhamnosidase Five different concentration of inoculum, 0.25%, 0.5%, 0.75%, 1% and 1.25% (v/v) of the standard stock inoculum was added in the juice and incubated for 24 h at room temperature. A differential response in rhamnosidase activity was obtained which showed that the 0.75% inoculum concentration exhibited maximum enzyme activity i.e. 0.057 IU mL-1 and 0.25% exhibited minimum rhamnosidase activity i.e. 0.023 IU mL-1 in kinnow juice (Fig. 1). Increase in inoculum size resulted in lesser enzyme production, due to the nutrient exhaustion and oxygen limitation. Similar results were also reported in Bacillus methylotrophicus (Mukund et al., 2014) and Staphylococcus xylosus MAK2 (Puri and Kalra, 2005) for naringinase production. Puri et al., (2005) studied the inoculum level of 3-15% (v/v) in the salt medium with naringenin as an inducer to establish the effect of inoculum size on the naringinase production by A. niger. They observed that 10% (v/v) inoculum was optimal for growth as well as naringinase production and the lag phase was also minimal. Effect of brix (°B) on α-L-rhamnosidase production In juice, brix was adjusted to 12, 13, 14, 15 and 16 °B by adding sugar solution followed by inoculation of yeast i.e. 0.75% (v/v). It was incubated for 24 h at room temperature. The effect of different °B on yeast rhamnosidase activity was tested and best °B for maximum rhamnosidase activity (0.05 IUmL-1) was 13 in juice (Fig. 1). Further, with the increase in the initial sucrose concentration the rhamnosidase production was decreased which indicated that the higher sucrose concentration had an adverse effect on the enzyme production efficiency of the yeast. Naringinase activity was repressed by glucose, sucrose, citrate and lactose although these carbon sources supported excellent growth (Puri et al., 2005; Bram and Solomons, 1965). Production of α-L-rhamnosidase by A. nidulans is mediated by carbon catabolite repression, which appears to be CreA-independent (Orejas et al., 1999). Further, it has been reported that the enzyme was not produced when A. kawachii was grown on 0.5% glucose as the sole carbon source (Koseki et al., 2008). Effect of incubation time on α-L-rhamnosidase production Effe


INTRODUCTION
India is the second largest producer of fruits, with a production of 44.04 million tonnes of fruits from an area of 3.72 million hectares and holds third rank in respect of production of citrus fruits in the world.Kinnow, a hybrid of Citrus nobilis and Citrus delicosa is a prevalent citrus fruit in Punjab covering an area of 46,000 hectares with the production of 9.88 lakh tones NHM (2014).
Kinnow mandarin juice has high therapeutic value as antispasmodic, sedative, cytophylactic, digestive, anti carcinogenic, anti inflammatory and anti allergic.The health benefits of citrus fruit juices have been attributed due to the presence of bioactive and antioxidant compounds.A total of 150 g edible portion of orange provides 0.3 g fiber and 17 g of carbohydrates that can supply upto 73 kilocalories.
Kinnow juice turns bitter after extraction due to chemical naringin (flavanoid) and limonin (limonoid).Naringin is the major component in citrus fruit with very bitter taste and a threshold of 20 mgKg -1 in water and detectable limit less than 1.5 mgKg -1 (Chen et al., 2010).The presence of limonin and naringin in excess of 6 ppm and 600 ppm respectively has been established as an objectionable level of bitterness in processed citrus products such as juice, wine and vinegar (Guadagni et al., 1973).Numerous techniques are used to reduce naringin such as adsorptive debittering (Fayoux et al., 2007), enzymatic hydrolysis (Puri and Kalra, 2005), poly-styrene divinyl benzene styrene resin treatment and β-cyclodextrin treatment (Mongkolkul et al., 2006).These techniques have limitations in altering nutrient composition either through chemical reactions or removal of nutrients, flavor and color etc.Another suitable debittering procedure is the stepwise hydrolysis of naringin by naringinase (Chen et al., 2010).The enzyme naringinase is composed of α-L-rhamnosidase (EC 3.2.1.40)and β-D-glucosidase (EC 3.2.1.21).Naringin (4′,-5,7′-trihydroxyflavonone-7-rhamnoglucoside) is first hydrolyzed by α-L-rhamnosidase activity of naringinase to rhamnose and prunin (one third of the bitterness of naringin) which can be further hydrolyzed into glucose and naringenin by the β-D-glucosidase component of naringinase.The potential application of rhamnosidase is used in the debittering of citrus fruit juices (Busto et al., 2007), manufacture of prunin from naringin, manufacture of L-rhamnose by hydrolysis of natural glycosides containing terminal L-rhamnose, enhancement of wine aromas by enzymatic hydrolysis of terpenyl glycosides containing L-rhamnose, elimination of hesperidin crystals from orange juices, conversion of chloropolysporin B to chloropolysporin C, the derhamnosylation of many L-rhamnose containing steroids for example, diosgene, desglucoruscin, ginsenosides-Rg2, etc. whose derhamnosylated products have their clinical importance (Feng et al., 2005).
The nutritional and therapeutic value of kinnow provides ample scope for processing into a value added fermented product with retention of organoleptic properties, nutritional attributes, characteristics sensory properties, flavor, aroma, texture and long shelf life.So, the aim of this work is to optimize and produce debittered kinnow beverage using α-L-rhamnosidase producing yeast Clavispora lusitaniae KF633446.

Yeast culture
Yeast strain producing rhamnosidase enzyme was isolated from whey beverage and identified as Clavispora lusitaniae (accession number-KF633446) on the basis of morphological, biochemical and 18S rDNA sequence analysis.
α-L-Rhamnosidase enzyme assay 50 mL of the diluted kinnow juice in Erlenmeyer flask (100 mL) was aerobically cultured at 30±5 °C for 48 h on a rotary shaker (150 rpm).After centrifugation (12,000 × g for 10 min), the supernatant was collected for measurement of rhamnosidase activity.The α-L-rhamnosidase activity (RA) was determined using p-nitrophenyl-α-L-rhamnoside (p-NPR, Sigma) as the substrate (Romero et al., 1985).The reaction mixture consisted of 0.1 mL of 4.8 mM p-NPR solution, plus 0.19 mL of 50 mM citric acid/Na citrate buffer, pH 5.0 and 10 µl of enzyme or buffer (for the blank) and was incubated at 50 °C.Aliquots of 50 µL from the reaction mixture were removed every 2 min and placed into 1.5 mL of 0.5 M NaOH.These aliquots were kept in an ice bath until the absorbance was measured at 400 nm (Rajal et al., 2009).One unit (U) of enzyme activity was defined as the amount of enzyme required to release 1 μmol of p-nitrophenol per minute.

Preparation of debittered kinnow beverage
A debittered kinnow beverage was prepared under optimized conditions of inoculum concentration, TSS, temperature and incubation time.

Extraction of juice
Kinnow (Citrus reticulata Blanco) was procured from Department of Fruit Science, PAU, Ludhiana, Punjab, India.Fruits were washed in chlorinated water and then used for the extraction of juice.Juice was extracted aseptically under hygienic conditions by kinnow juice extractor.

Preparation of sugar solution
The sugar solution was prepared by boiling (500 g) granulated sucrose in one litre of water for 10 min and then allowed to cool at room temperature and stored aseptically in sterilized glass bottles.

Inoculum preparation
The inoculum was prepared in diluted juice with brix adjusted to (13°B).A loopful culture of 24 h old yeast (Clavispora lusitaniae KF633446) was inoculated in 100 mL diluted kinnow juice in 250 mL Erlenmeyer flask and incubated at 30±5 °C for 24 h to achieve concentration of 10 5 -10 6 cells mL -1 .

Fermentation, bottling and storage
The physico-chemical analysis (pH, % acidity, TSS, brix acid ratio, naringin, limonin and juice yield) of fresh kinnow juice was performed.Juice was diluted in the ratio 1:1.5 with water.Diluted juice was pasteurized at 82 °C for 15 sec, cooled and brix adjusted to 13 °B by adding sugar solution followed by inoculation of yeast i.e. 0.75% (v/v).It was incubated for 48 h at 30±5 °C.The beverage was refrigerated for 24 h, siphoned, bottled and stored in refrigerated conditions.

Shelf life determination of kinnow beverage
Shelf life fermented debittered kinnow beverage, stored at refrigerated temperature (4 °C) was studied and evaluated fortnightly for physicochemical, microbiological and sensory qualities.

Physiochemical and microbiological analysis of kinnow juice and beverage
The total soluble solids and pH of kinnow juice and beverage were determined by using Erma hand refractometer of 0-32 °B (Erma, Tokyo, Japan) and pH meter (ECIL, Hyderabad, type 101; Electronic Corporarion of India Ltd., Hyderbad, India).Total acidity expressed as citric acid was estimated following the procedure of AOAC (1999).Brix-acid ratio was calculated through dividing TSS value by total acidity of the juice and carbonated beverage.Total sugars were estimated by phenol sulphuric acid method (Dubois et al., 1956).Reducing sugars were estimated by the method of Miller (1959).The titration method using 2, 6-dichlorophenol indophenol dye was used to estimate ascorbic acid (AOVC 1996).The total phenolic content (TPC) was determined by spectrophotometry, using gallic acid as a standard, according the method described by Singleton and Rossi (1965).Limonin content was estimated by colorimetric method (Vaks and Lifshitz, 1981) and naringin content was estimated by Davis method (1947).Carbon dioxide volumes in beverage bottles were determined by Zahm and Nagel piercing device (CO 2 tester, Zahm and Nagel Co., Inc., Holland, New York, USA) and percent alcohol (v/v) was calculated by spectrophotometric determination method of ethanol (Caputi et al., 1968).Total yeast count was enumerated on GYE agar by serial plate dilution method.

Sensory evaluation
The organoleptic evaluation of kinnow beverages was done on the basis of appearance, taste, color, aroma, bouquet, body, flavor, astringency and overall acceptability by a panel of judges.Consumer acceptance for the products was evaluated on a nine point "Hedonic scale" (Amerine et al., 1965).

Statistical analysis
Statistical analysis was done by using CPCS1 software.Standard errors were calculated for all mean values.Differences were considered significant at the p ≤ 0.05 level.

Screening of juice components for optimization of α-L-rhamnosidase production Effect of percent inoculum concentration on α-Lrhamnosidase
Five different concentration of inoculum, 0.25%, 0.5%, 0.75%, 1% and 1.25% (v/v) of the standard stock inoculum was added in the juice and incubated for 24 h at room temperature.A differential response in rhamnosidase activity was obtained which showed that the 0.75% inoculum concentration exhibited maximum enzyme activity i.e. 0.057 IU mL -1 and 0.25% exhibited minimum rhamnosidase activity i.e. 0.023 IU mL -1 in kinnow juice (Fig. 1).Increase in inoculum size resulted in lesser enzyme production, due to the nutrient exhaustion and oxygen limitation.Similar results were also reported in Bacillus methylotrophicus (Mukund et al., 2014) and Staphylococcus xylosus MAK2 (Puri and Kalra, 2005) for naringinase production.Puri et al., (2005) studied the inoculum level of 3-15% (v/v) in the salt medium with naringenin as an inducer to establish the effect of inoculum size on the naringinase production by A. niger.They observed that 10% (v/v) inoculum was optimal for growth as well as naringinase production and the lag phase was also minimal.

Effect of brix (°B) on α-L-rhamnosidase production
In juice, brix was adjusted to 12, 13, 14, 15 and 16 °B by adding sugar solution followed by inoculation of yeast i.e. 0.75% (v/v).It was incubated for 24 h at room temperature.The effect of different °B on yeast rhamnosidase activity was tested and best °B for maximum rhamnosidase activity (0.05 IUmL -1 ) was 13 in juice (Fig. 1).Further, with the increase in the initial sucrose concentration the rhamnosidase production was decreased which indicated that the higher sucrose concentration had an adverse effect on the enzyme production efficiency of the yeast.Naringinase activity was repressed by glucose, sucrose, citrate and lactose although these carbon sources supported excellent growth (Puri et al., 2005;Bram and Solomons, 1965).Production of α-L-rhamnosidase by A. nidulans is mediated by carbon catabolite repression, which appears to be CreA-independent (Orejas et al., 1999).Further, it has been reported that the enzyme was not produced when A. kawachii was grown on 0.5% glucose as the sole carbon source (Koseki et al., 2008).

Effect of incubation time on α-L-rhamnosidase production
Effect of different incubation time on enzyme activity was studied.Kinnow juice (brix 13 °B and inoculum concentration 0.75% v/v) was incubated for different time periods (12, 24, 36, 48 and 60 h) at room temperature.Fig. 1 shows that the maximum enzyme activity in juice was observed after 48 h of fermentation.Maximum naringinase production (12.05UL -1 ) was observed at 34 h of fermentation, which corresponds to a stationary phase of growth (Mukund et al., 2014).In the batch reactor, the maximum α-rhamnosidase activity was obtained after 10 days from Penicillium ulaiense (Rajal et al., 2009).The reduction in production time is important because it decreases the fermentation costs and contamination with opportunistic microorganisms in scale up process.

Effect of temperature on α-L-rhamnosidase production
Five different levels of temperature were also studied.Kinnow juice (brix 13 °B and inoculum concentration 0.75% v/v) was incubated at different temperatures (15, 20, 25, 30 and 35 °C) for 48 h.Data in graph in Fig. 1 show the effect of changing temperature on enzyme activity.As the temperature increases, initially the enzyme activity increased while it decreased at higher temperatures at the same time.The decrease in enzyme activity may be the deactivation of enzyme due to the weakening of non covalent interactions that stabilize the protein structure.The optimum temperature for Pichia angusta rhamnosidase was observed at 40 °C (Yanai and Sato, 2000).The reported temperature optima of α-L-rhamnosidases are in the range of 40-80 °C though one bacterial α-L-rhamnosidase active at 4 °C is reported (Orrillo et al., 2007).

Scale up of the optimized process in the laboratory bench scale fermenter
The optimized process parameters (brix 13 °B, inoculum concentration 0.75% v/v, temperature 30 °C and incubation time 48 h) for production of maximum rhamnosidase enzyme was used for preparation of fermented debittered beverage.The experiment was conducted in laboratory fermenter (capacity-10L) at Department of Microbiology.

Evaluation of microbiological and physicochemical properties of kinnow beverage during storage
The results of microbiological and physicochemical properties of kinnow beverage during storage period of 90 days are summarized in Table 2.The results showed a significant decrease in brix from 13.00±0.20°B to 11.20±0.30°B and brix acid ratio from 92.85±0.00 to 19.31±0.00.Similar results have been reported by Sarolia and Mukherjee, 2002 in their studies on lime juice, Khandelwal et al., 2006 during the fermentation of kinnow sera, cane and kinnow cane juice and Ahmed et al., 2008 in preparation of ready to serve mandarin (Citrus reticulata) diet drink.The increase in TSS content of juice during storage might be due to hydrolysis of polysaccharides into monosaccharide and oligosaccharides.
The pH also decreased from 3.40±0.10to 3.00±0.10,while acidity increased from 0.14±0.03% to 0.56±0.01%after fermentation.pH is inversely proportional to the acidity of any medium.This decrease in pH and increase in acidity was attributed to formation of acidic compounds by degradation of reducing sugars (Zia, 1987;Akhtar et al., 2010).A similar result of decreasing pH was also reported by Saleem, 1980 andAhmed, 2008.
The polyphenol contents of commercial fruit juices in the case of pineapple, orange and mango juices were higher than those of Thai beverages, reported by Abdullakasim et al., (2007).Different factors such as processing techniques, clarification and pasteurization can affect polyphenol contents of commercial juices.
According to Ritter et al. (1992) and Karadeniz and Eksi (2001) reports, clarification also decrease the polyphenolic contents of commercial fruit juices.
Polyphenol contents decreases constantly with the progress of the ripening, while in red coloured varieties it increases during the last ripening stage due to the maximal accumulation of anthocyanidines and flavonols (Marinova et al., 2005).
The decrease of limonin from 6.90±0.10 to 3.52±0.10ppm might be due to production of CO 2 during storage.Carbon dioxide at pressures of 21 to 41 MPA at 30°C-60°C for 1h resulted in an average removal of 25% of the limonin from navel orange juice (Kimball, 1987).A gradual increase in limonin content in juice blends with storage period might be due to conversion of a chemical compound limonate-a-ring lactone (non-bitter) in to limonin (bitter) in the juice (Premi et al., 1994).The decrease of naringin with storage was 443.00±10.00 to 143.70±4.00ppm due to hydrolysis of naringin into rhamnose and prunin by α-L-rhamnosidase activity of yeast.The alcohol production starts after 10 days and gradually increased from 0.11±0.01%to 0.89±0.05%after 90 days.The CO 2 pressure 0.65±0.05bar starts building after 10 days and reached up to 1.46±0.06bar after 90 days.Sensitivity of yeast cells to ethanol marginally increased on decreasing the pH from 6.00-3.00.During fermentation process, CO 2, alcohol and glycerol produced is proportional to the amount of sugar fermented.
The yeast strain produced large amount of glycerol at the expense of ethanol represent an advantageous alternative for the development of beverages with low ethanol content versus physical process which alter the organoleptic properties of the final product (Jairath et al., 2012).Total yeast count was increased from 58x10 5 ±10.00 to 58x10 8 ±20.00 cfu mL -1 at the end of 90 days.This study indicated that the shelf life of beverage was 90 days.

Evaluation of sensory attributes of beverage during storage
The changes in sensory attributes like taste, color, aroma, bouquet, flavor and astringency of kinnow beverage were analyzed once every 10 days.All the sensory parameters were stable at storage period (90 days) with almost no change in organoleptic sensation (Table 3).Beverage was found to be acceptable up to 3 months of storage.
The storage temperature can greatly affect the beverage tastes and smells.Lower temperatures will emphasize acidity and tannins while muting the aromatics.Higher temperatures will minimize acidity and tannins while increasing the aromatics.The presence of yeast in beverage gave a desirable freshness to the fermented beverage due to production of carbon dioxide and ethanol.Bhardwaj (2013) also reported that the low temperature and high relative humidity did not cause any change in qualitative characters and palatability of stored juice and helped in maintaining juice flavor, colour, TSS: acid ratio and sugars in balanced form than the ambient storage condition.In ambient condition change in colour of kinnow juice might be attributed to oxidation of phenolic compounds present in juice and maillard reaction between sugars and amino acids (Gonzalez, 2000).A gradual decrese in flavour and taste which may be due to the degradation of ascorbic acid and furfural production (Kausar et al., 2012) and may also be due to heat treatment applied during processing (Pruthi et al., 1984).

CONCLUSION
On the basis of results it can be concluded that the strain Clavispora lusitaniae is capable for producing debittered kinnow beverage using the optimized process parameters.
The beverage can be stored for a period of 3 months at refrigeration temperature with minimum changes of all physico-chemical characters.Thus the technology presented here can also redress the problem of bitterness in food industries by reducing the naringin content of citrus juices.
in planning and writing of experiment; optimization of parameters for beverage preparation.